Sunday, May 14, 2006

Ian Chappell bats for Bucknor

A great cricketer he may be, but Ian Chappell, writing here while arguing the use of technology, gets simple logic mixed up
If you want a clue that the ICC’s current philosophy on technology is muddled then look no further than the response of former South African keeper Dave Richardson to Bucknor’s criticism. The ICC’s general manager — cricket, says about Bucknor’s umpiring year so far, “He’s averaging 96 % of his decisions being correct.”

I assume, to a large degree this is in the opinion of technological aids like Hawk-Eye and Snickometer.

However, in the words of Ten Sports (Dubai) head of production Steve Norris, “The Hawk-Eye is 90 % accurate, that is what they (the creators) claim.”

Excuse me; “I’m sorry I’ll read that again.” An umpire is getting 96% of his decisions right, while on the other hand Hawk-Eye is guaranteed 90% correct. So, why do we need technology to protect the umpires?
As I've said before, it doesn't really take a rocket scientist to figure out that the "96%" statistics hailed as the good performance by Bucknor includes the straightfoward chances as well. I'm more than sure his average for contentious decisions would be lower than that.

As for 90% accuracy of Hawk-eye, two points. Firstly, no-one is using hawkeye to actuallly adjudicate on decisions - even the ICC's new ruling specifically rules out hawkeye and snickometer for which, btw, no-one told us the accuracy percentage. I have a suspiscion it may exceed the Bucknor barometer!

The second point is that the statement from Steve Norris is a blanket one. We don't know if Hawk-eye is less accurate in certain specific cases than others. But being a computer program, I would guess its flaw may be more evenly distributed.

2 Comments:

Anonymous Anonymous said...

This is really interesting. We need to know how this 96% figure is made up. Nowadays players just appeal for any random thing and if you count that in the 96% then i am sure a globus monkey could get 96%.

90% accuracy for hawk-eye seems a low figure. I read elsewhere that it was much higher than that, though i don't have the link at hand. But i think the ICC should publicly release the methodology they apply at arriving at the 96% along with the way the marked particular appeals and then fans could go through and argue if they truly evened out in the end.

Ram

May 14, 2006 11:47 PM  
Blogger worma said...

It would be great if you can find that link, Ram. I believe Norris would have given the most cautious figures there, and even I seem to remember reading somewhere that hawkeye technology was used in defence equipment...for which 90% would surely be low.

Anyway, as you said, the 96% number for Bucknor surely includes all those simple calls like plumb LBW, caught in the deep, clear caugh behind, LBW calls for ball pitching well outside leg etc. People like Richardson try to hide behind numbers by using it this way. I hate it!

May 15, 2006 10:35 AM  

Post a Comment

<< Home